Earth observation models with spatial validation
Land cover, building and damage models are trained and tested on spatially separated blocks, so reported accuracy reflects new places rather than the neighbours of training pixels.
GCRI · Foundry
AI and machine learning at Foundry
Artificial intelligence and machine learning extend detection, classification and forecasting across large archives, and agents extend them into action. In this field the method is also a source of risk: models that fail when conditions shift, outputs that sound certain when they are not, and tools that turn an error into an effect.
01Contents
AI and machine learning at Foundry in five parts.
02Application
What artificial intelligence and machine learning does here.
Artificial intelligence and machine learning develop algorithms that learn patterns from data to classify, predict or generate outputs. They are increasingly used for hazard detection, forecasting and decision support, and are themselves a source of risk. In data and compute and methods and modelling, machine learning can scale detection and forecasting across interconnected systems. Its own failure modes must be weighed when automated outputs inform consequential decisions. The canonical description is on the STEM Commons.
Model cards and promotion records
Each model version carries its intended use, training data, evaluation by region and subgroup, known limits and red-team results, and is promoted only with a recorded review.
Scoped agents with receipts
Agents run analyses with a declared tool set, data classes, budget and kill-switch enforced at the policy gate, and every action produces a receipt for a reviewer.
03Monitor
Model training.
The screen a team would use to follow this method: its measures, states and decisions. Names are examples and values are illustrative.
Model training
Foundry · AI and machine learning at Foundry · Run R-6794
Drift on one sliceGCRI · Foundry
Evaluation by slice
| All | 0.83 | 0.02 | 5712 | Pass |
| Region A | 0.74 | 0.05 | 1098 | Review |
| Region B | 0.76 | 0.05 | 458 | Review |
| Rare class | 0.64 | 0.07 | 160 | Below bar |
Method card
- Method
- Artificial intelligence and machine learning
- Split
- time-based, no leakage
- Features
- 22
- Label
- MODELLED
Population stability
04Data
What the method consumes.
Datasets and records used by artificial intelligence and machine learning on this platform, with the role each plays and the label its values carry. Sources are listed in the source register.
| Data | Role | Label |
|---|---|---|
| Labelled training sets | Supervised learning | REPORTED |
| Satellite image archives | Inputs to classification | OBSERVED |
| Held-out and out-of-distribution test sets | Evaluation | OBSERVED |
| Evaluation and red-team records | Promotion decisions | REPORTED |
| Datasheets for training data | Origin, consent and licence | DECLARED |
05Methods and evidence
How AI and machine learning at Foundry is modelled, measured and tested.
The model, its assumptions, its data and its uncertainty, in the form a scientist or engineer would check. Structures follow published methods. The values drawn are illustrative, not findings.
Model, equations and assumptions
The model structure used for AI and machine learning at Foundry, the assumptions it makes and how each is tested.
Risk of element i
Ri = f(Hi, Ei, Vi, Ci)hazard, exposure, vulnerability and coping capacityExpected annual loss
EAL = ∑k pk · Lkover events k with annual probability pkCoupled systems
xt+1 = A xt + B ut + εtA carries the couplings, u the shock, ε the error| Assumption | Why it is made | How it is tested |
|---|---|---|
| Hazard and exposure are independent within an event | Keeps the loss integral tractable | Compare with a copula fit on joint records |
| Couplings are linear near the operating point | Lets shocks propagate through A | Perturb beyond the linear range and compare |
| Vulnerability curves transfer between places | Local curves are rarely available | Validate against observed losses where they exist |
- Inputs
- versioned datasets with provenance
- Outputs
- distributions, never single numbers
- Code
- released with the model card and tests
- Reproduction
- same data, code, environment and seed
FAIR guiding principles · GO FAIR and FORCE11W3C PROV · World Wide Web ConsortiumGuidance on consistent treatment of uncertainties · IPCCCEOS Analysis Ready Data · Committee on Earth Observation Satellites
06Validation
How results are checked.
Models are evaluated on held-out data that respects spatial, temporal and group structure. Robustness, calibration and fairness are tested alongside accuracy.
| Check | How it is done |
|---|---|
| Spatial and temporal hold-out | Test data are separated in space and time from training data to prevent leakage |
| Probability calibration | Predicted probabilities are compared with observed frequencies on held-out data |
| Drift monitoring | Input and output distributions in use are compared with those at validation, and divergence triggers review |
| Adversarial testing | Models and agents are tested against prompt injection and misuse before promotion |
Performance degrades when deployment conditions differ from training data. Models can encode biases and fail without clear warning on rare events.
07Process
How the process runs.
Process
AI and machine learning at Foundry: AI risk management (NIST)
The AI RMF Core has four functions, GOVERN, MAP, MEASURE and MANAGE, with GOVERN cross-cutting; the order drawn is one common reading, not a checklist.
01 Govern
- 01GOVERN: policies, roles and accountabilityStep · Model stewardGOVERN is a cross-cutting function. It is infused throughout AI risk management and enables the other functions.
02 Map
- 02MAP: context and intended useDatasetThe MAP function establishes the context to frame risks related to an AI system, including intended purposes and impacts on people.
03 Measure
- 03MEASURE: evaluate trustworthy characteristicsModelAI systems are evaluated for trustworthy characteristics. They should be tested before deployment and regularly in operation.
- 04MEASURE: track identified risks over timeModelMechanisms for tracking identified AI risks over time are in place, including unanticipated and emergent risks.
04 Manage
- 05MANAGE: treat risksStepRisk treatment plans cover response to, recovery from and communication about incidents. Resources go to mapped and measured risks.
- 06Proceed or not (MANAGE 1.1)Gate · Model stewardA determination is made whether the system meets its purposes and objectives, and whether to proceed. The framework sets no thresholds.
- 07Deployment decisionDecision · Model stewardThe organisation that owns the system decides whether and under what conditions to deploy.
- 08Post-deployment monitoring and incidentsModelPost-deployment monitoring plans are implemented. Incidents and errors are shared with relevant AI actors, including affected communities.
05 Record
- 09Inventory of AI systemsDatasetGOVERN 1.6 asks for mechanisms to inventory AI systems. A model card is common practice and is not named in the framework.
Neutral outline of a published external standard or method, from its issuing body. Not a record of operation. Nexus does not perform, certify or approve any step shown. A gate or decision belongs to the party the source names. The actions are not necessarily an ordered set of steps. The EU AI Act is not drawn in this flow.
08Connections
Topics that rely on this method.
09Stewardship
Who maintains the method.
Artificial Intelligence and Autonomous Systems stewards artificial intelligence and machine learning for the Consortium and reviews how it is applied across platforms.
10Connected
Where this page leads across the Consortium.
11Questions
Questions and answers.
What are the limits of artificial intelligence and machine learning in this field?
Performance degrades when deployment conditions differ from training data. Models can encode biases and fail without clear warning on rare events.
How are results checked?
Models are evaluated on held-out data that respects spatial, temporal and group structure. Robustness, calibration and fairness are tested alongside accuracy.
Are the values on this page real?
The data names, roles and checks are real descriptions of practice. The canvas shows a design of the method’s use, not a run; no result is reported here.
Where else is this method applied?
The canonical method page lists every platform that applies it.
ReportsContribute
Contribute evidence from this page.
Reports is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.
GCRIFoundryFoundry Build and Readiness
- EmailGateway GW-GCRISend it from this pageEvidence, a correction, a case, a method, a review, a translation or profile input. It arrives tagged with this page, so the editor who owns it receives it.
- ZenodoOpen depositDeposit it and submit to the communityPublish your own record on Zenodo with your ORCID iD, then submit it to the Reports community there. Curators review it; the DOI stays yours.
- ReportsGCRISee every way to contributeThe contribution types, how each is reviewed, what credit you receive, the licence and what is never accepted.
Never send personal data, sensitive infrastructure detail or anything you have no right to share. A contribution is reviewed before use; sending it confers no authorship or editorial rights, and sponsorship never buys a finding.
Foundry
Bring a method, a dataset or a question.
Information and enquiries are open today. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.
Legal
Legal documents.
Maps, data and figures on this site are shown for context and general information. They are not advice, forecasts or warnings; the limits of the information are set out in the Disclaimers and Data Notices.